Yendo . — On the Mttcilage Glands of Undaria. 617 
A piece of frond fixed beforehand in the picric acid and preserved in 
a strong alcohol is imbedded in paraffin. The sections are stained first in 
aniline blue. They are treated with 70 per cent, and 90 per cent, alcohol 
successively until the colouring matter is washed out without any shade of 
blue in the tissue except in the glands. The sections are then stained in 
fuchsin, then sealed in Canada balsam after the usual process. The glands 
are now stained a deep violet and the other parts of the tissue a beautiful red. 
Delafield’s haematoxylin also answers well. In it the glands are stained a 
deep purple, while the remaining parts take a bluish violet colour. The 
former method of staining gives a sharper contrast of colours, while the 
latter is better for the purpose of studying the plasmic contents of the 
tissues. 
The glands have their origin in the cortical cells which are situated 
directly beneath the epidermal layer (Fig. 8). These cortical cells are so 
closely in contact with the overlying cells as to give a distromatic 
appearance to the epidermal layer. That the epidermal layer is never 
distromatic may be proved by comparing the cell-contents and the future 
development of both layers. As the first step of the gland formation, one 
of the cortical cells begins to swell up gradually, undoubtedly gaining its 
nourishment from its contiguous cells. As the glandular cell swells up, the 
epidermal cell situated just above it is gradually compressed upwards. The 
surface of the latter retains its former position in the same level with the 
neighbouring ones. The consequence is the diminishing of the cell-cavity. 
The cell-contents are dissipated by degrees, but the nucleus is limited to 
the narrow space now allowed for it (Fig. 10). The glandular contents 
are at this stage already differentiated, so as to assume a deeper staining 
than the others, and change into a hyaline plastic mass filling up the whole 
cavity of the swollen cell. The epidermal cell is finally flattened into a 
thick hyaline membrane, roofing over the gland. Hence it may be clearly 
understood that there is not any actual perforation upon a gland. Thus 
the primary supraglandular space is formed. 
In the surface view of the epidermal layer we often find two or three 
contiguous cells in course of degeneration (Fig. 4). This may be interpreted 
in two ways. In the first case two or more cells are sharing in the 
formation of one supraglandular space, as shown in Fig. 9. In the 
second, two or more supraglandular spaces may be in course of formation, 
each with a gland below, as shown in Fig. 10. Practically, however, the 
former case seems to be more frequent than the latter. As an abnormal 
case it was found that the epidermal cell upon the embryonal glandular 
cell did not degenerate, but both fused together into a long cylindrical cell 
(Fig. 13). In a rare case three cells, the one epidermal and the others 
cortical, serially disposed perpendicularly to the surface of the frond, have 
swollen up as shown in Fig. 14. The innermost cell stained deepest, in 
